Ah, yes, I skipped a step. The polls are based on old voting patterns, in which the bureaucratic class is over-represented
-
-
This makes it all make much more sense. What you are terming the bureaucratic class is also many times more likely to do survey research/polls/etc compared to all other groups.
1 reply 0 retweets 3 likes -
Pollsters take this into account The polls were off by 2.8% in Trump's favour in 2016. In 2018, the polls were off by 0.2%.... in the Democrats' favour. In 2020 polls show biden 5-10 points ahead. If Trump wins, this is by far the worst polling error of all time.
1 reply 0 retweets 1 like -
To a point, I worked in social science research doing survey design and administration for awhile (I’m a statistician by trade) and the reality is you can’t correct as much as one might assume. In political polling it often doesn’t matter as the responders are also the voters
2 replies 0 retweets 2 likes -
They corrected very successfully in 2018; the error was 0.2% in favour of Democrats. Indeed, 2016 was just an over-correction from 2012, where they underestimated support for Democrats.
1 reply 0 retweets 1 like -
I’d love to read up on this, can you link to the data you are looking at?
2 replies 0 retweets 2 likes -
Nate Silver essayhttps://www.google.co.uk/amp/s/fivethirtyeight.com/features/the-polls-are-all-right/amp/ …
2 replies 0 retweets 1 like -
LOL their defense is "polls were never accurate!" - I knew that!
2 replies 0 retweets 3 likes -
Polls always have a margin of error. But look at that 1980 error!
1 reply 0 retweets 1 like -
I see a significant source of error this time too. The number of people voting is something like double what it was last year. No one knows how these new voters will vote
2 replies 0 retweets 3 likes
Exactly
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.